如何从 python 的列表中消除具有某些字符或短语的元素？

Question

我有一个来自 excel 电子表格的植物名称列表，我用 pandas 提取了该电子表格。 在删除重复项并将整个列表设为小写之后，我想删除括号、撇号、破折号等字符以及“A”和“The”等短语，以进一步消除任何可能的重复项，以便在如下列表中：（ ("A Pumpkin", "Pumpkin", "The Pumpkin", "Pump-kin", "(European) Pumpkin", "Pumpkin (Orange)", "Farmer's Pumpkin")仅保留“Pumpkin”。 注意我不想只从字符串中删除字符，而是从列表中删除整个字符串。

def checkSyntax(str):
    boolean = None

    regexes = ["a ", "the ", "^\W"]
    combined = "(" + ")|(".join(regexes) + ")"
    match = re.match('combined', str)
    if match == None:
        boolean = True

    return boolean

def elimInvalidNames(names):
    new_names = [s for s in names if checkSyntax(s)]
    return new_names 

test_list = ("A Pumpkin", "Pumpkin", "The Pumpkin", "Pump-kin", "(European) Pumpkin", "Pumpkin (Orange)", "Farmer's Pumpkin")
test_list = list(map(lambda x: x.lower(), test_list))
test_list = elimInvalidNames(test_list)
print(test_list)

出于某种原因，这去掉了“the”和“a”，但没有去掉括号、破折号或撇号。

Answer 1

这应该这样做：

import re
new = []
test_list = ("A Pumpkin", "Pumpkin", "The Pumpkin", "Pump-kin", "(European) Pumpkin", "Pumpkin (Orange)", "Farmer's Pumpkin")
for s in test_list:
    for n in s.split():
        if n == re.sub(r'[^\w\s]','',n) and n.lower() != 'a' and n.lower() != 'the': # Adds word to new list if word is not 'a', 'the', and doesn't contain punctuations 
            new.append(n)
print(list(set(new))) # Convert to a set to remove duplicates, then back to alis

Output：

['Pumpkin']

Answer 2

如果不强制使用正则表达式，请尝试：

names = ("A Pumpkin", "Pumpkin", "The Pumpkin", "Pump-kin", "(European) Pumpkin", "Pumpkin (Orange)", "Farmer's Pumpkin")
def checkSyntax(s):
    regexes = ["a ", "the ",'"', "'", '\\', '.', ',', ')', '(', '-']
    return not any(letra in s.lower() for letra in regexes)

def eIn(names):
    new_names = [s for s in names if checkSyntax(s)]
    return new_names

Answer 3

如果要使用正则表达式，请使用re.search ，因为这将尝试匹配字符串任何部分的表达式。 re.match function 仅尝试匹配字符串开头的表达式。

例如，以下代码将列表过滤为['Pumpkin'] ：

import re
invalid_names_re = re.compile(r"(A )|(The )|[()\-']", re.IGNORECASE)                                                                                                                                                            names = ("A Pumpkin", "Pumpkin", "The Pumpkin", "Pump-kin", "(European) Pumpkin", "Pumpkin (Orange)", "Farmer's Pumpkin")  

filtered_names = [name for name in names if not invalid_names_re.search(name)]